Professional Documents
Culture Documents
Ulm University
1 Introduction
With increasing frequency the length of electromagnetic waves decreases. If the wavelength
is equal or smaller than the length of the transmission line, voltage and current can not be
assumed as constant over the length of the line. It has to be considered, that voltage and
current depend on the z coordinate. In Fig. 1 you see a two-wire line and in Fig. 2 the
local voltage curves at three dierent time points. TEM means transversal electromagnetic,
consequently the wave has only transversal eld components.
z
Fig. 1: Equivalent circuit of an ideal TEM transmission line.
u(z) t0
t0 + ∆t
t0 + 2∆t
voltage and current waves move along the line with the time.
The theoretical basics for dening the wavelength on the transmission line and the correlation
between current and voltage is the Transmission Line Theory and subject of the next chapter.
2 Theoretical Basics
To cover all voltage and current distributions on transmission lines, we have to solve the
Maxwell equations including the boundary conditions. The example below of a two-wire
transmission line, however, can be described as an innitesimal piece of transmission line as
shown in the equivalent circuit diagram 3. Consequently, a nitely long transmission line exists
from an unnite number of those innitesimal pieces.
u(z + ∆z, t) =
u(z, t) G0 ∆z C 0 ∆z u(z, t) + ∆u(t)
∆z
uR = R0 ∆z iR , (1)
∂iL
uL = L0 ∆z , (2)
∂t
iG = G0 ∆z uG , (3)
∂uC
iC = C 0 ∆z . (4)
∂t
Using the mesh rule (KVL=Kirchho´s voltage law):
∂iL
u = R0 ∆z iR + L0 ∆z + u + ∆u . (5)
∂t
Using the node analysis (KCL=Kirchho´s current law):
∂(u + ∆u)
i = G0 ∆z(u + ∆u) + C 0 ∆z + i + ∆i . (6)
∂t
Neglecting dierences of second order the following equation for the current in valid:
∂u
i = G0 ∆z u + C 0 ∆z + i + ∆i . (7)
∂t
If (5) and (7) are divided by ∆z , in the limit ∆z→∂z ; that means the dierence equations
become dierential equations:
∂u ∂
= − R0 + L0 i, (8)
∂z ∂t
∂i 0 0 ∂
=− G +C u. (9)
∂z ∂t
These are the dierential equations (DE) of the electrical transmission line forming a system
of partial DEs of rst order.
In the following we focus on the steady state, that means we deal only with a particular
solution of the DEs with sine-shaped excitation:
dI
= jωI , (10)
dt
dU
= jωU . (11)
dt
With the help of these transformation equations (transition to the complex AC calculation ) it
follows from the DEs (8) and (9):
dU
= −(R0 + jωL0 )I , (12)
dz
dI
= −(G0 + jωC 0 )U . (13)
dz
These are the dierential equations of electric transmission lines in steady state ; it is a linear
system of normal DEs of rst order with constant coecients. For the solution of this system
of equations we dierentiate (12) with respect to z and plug this in (13). So you get the wave
equation of this transmission line, the so-called telegraphic equation :
2
d U 2
2 =γ U, (14)
dz
p
with γ = (R0 + jωL0 )(G0 + jωC 0 ) = α + jβ . (15)
γ : propagation constant
2
d I 2
2 =γ I. (16)
dz
One possible set-up for the solution of the telegraphic equation (14) is:
1 dU γ
−γz γz
I=− = U h0 e − U r0 e . (18)
R0 + jωL0 dz R0 + jωL0
s
R0 + jωL0 R0 + jωL0
Z0 = = (19)
γ G0 + jωC 0
results in:
1
I(z) = U h0 e−γz − U r0 eγz = I h0 e−γz − I r0 eγz . (20)
Z0
Thus it follows:
Uh U
= − r =: Z 0 . (21)
Ih Ir
Z0 is called characteristic impedance. This is the impedance which the wave faces when
travelling in positive z direction on an unnitely long transmission line.
The characteristic impedance is assumed to be real in most practical cases. In this experiment,
too, we consider this parameter as real.
then the values of voltage and current are determined for each z:
1
I(z =0) = Ia = (Uh0 − Ur0 ) , (24)
Z0
Ia Z0 = Uh0 − Ur0 , (25)
1
(23)−(25): Ur0 = (Ua − Ia Z0 ) , (26)
2
1
(23)+(25): Uh0 = (Ua + Ia Z0 ) , (27)
2
1 1
(17) : U = (Ua + Ia Z0 )e−γz + (Ua − Ia Z0 )e γz , (28)
2 2
1 1
(18) : I= (Ua + Ia Z0 )e−γz − (Ua − Ia Z0 )e γz , (29)
2Z0 2Z0
U = U a cosh(γz) − Z0 I a sinh(γz) , (30)
Ua
I = I a cosh(γz) − sinh(γz) . (31)
Z0
With
Ua
= ZL (33)
Ia
we get
Z L cosh(γz) − Z0 sinh(γz)
Z(z) = Z0 (34)
Z0 cosh(γz) − Z L sinh(γz)
or
Z L − Z0 tanh(γz)
Z(z) = Z0 . (35)
Z0 − Z L tanh(γz)
In case of a loss less transmission line (the losses were already assumed to be that low that
the characteristic impedance Z0 was assumed as real) we get
γ = jβ , (36)
Z L − jZ0 tan(βz)
=⇒ Z(z) = Z0 . (37)
Z0 − jZ L tan(βz)
In the following we deal with the problem of nding out the terminating impedance ZL from
the voltage characteristic of a TEM transmission line.
R 0 = G0 = 0 , (38)
√
and with this: α = 0, γ = j β = j ω L0 C 0 . (39)
For the discussion on the relations of an incident and reected wave we dene the below
parameters:
Ur I
reection coecient: r= =− r , (40)
Uh Ih
s = max ,
U
(V)SWR: (41)
U min
1
inverse (V)SWR: m= . (42)
s
The abbreviation (V)SWR stands for (voltage) standing-wave ratio. Furthermore, the follow-
ing equations are valid:
s−1
|r| = , (43)
s+1
1 + |r|
s= . (44)
1 − |r|
!
2γz −γz
U = U h0 1 + r0 e e , (45)
!
U −γz
I = h0 1 − r0 e2γz e . (46)
Z0
U (z =0) U
(45),(46) : = a = ZL (47)
I(z =0) Ia
ZL 1 + r0
we get: = (48)
Z0 1 − r0
Z L − Z0
⇐⇒ r0 = . (49)
Z L + Z0
Three frequently appearing special cases are:
Matching means that a transmission line having a wave impedance Z0 is connected with exactly
the same impedance.
Maximum voltages exists at the points where incident and reected wave are in-phase and
their magnitudes add up (constructive interference):
r0 = |r0 | e jφ . (51)
φ nπ φ λ
or: zmax = − − =− −n n = 0, ±1, ±2, . . . (53)
2β β 2β 2
λ
Thus, two neighboured maximum voltages are by
2 apart.
Analogue calculation for the minimum voltages yields
φ (2n + 1)π φ λ
zmin = − − =− − (2n + 1) n = 0, ±1, ±2, . . . (54)
2β 2β 2β 4
UE (z)
a) UE+,max
b)
UE+,min
−2π −π 0
βz
In Fig. 5 the voltage distribution on a transmission line with with a short circuit at the end of
the transmission line, curve a), Z L =0, and for a complex load impedance, curve b), is shown.
The voltage distribution means the envelope, the curve of the maximal and minimal voltage
UE (z) (or current IE (z)) which occurs over time. In the gure only the positive voltage is
depicted, since the curve of the negative voltage is only the mirror image on the abscissa and do
not contain more information. The envelope for both curves can be expressed mathematically
as:
UE+ (z) = max Û (z) cos(ωt + φ) = Û (z) , (55)
UE− (z) = min Û (z) cos(ωt + φ) = −Û (z) . (56)
The envelope of the current is similar to the curve of the voltage but the course is shifted by
λ/4 with respect to the course of the voltage.
In the case the transmission line was connected to a load not able to convert active power
like short-circuit, open-end, capacitor or inductor, the amplitude of the reected wave has the
same value as that of the incident wave. The destructive interference results in points where
the voltage is zero, as shown in curve a). In the case of a real or complex load, the reected
amplitude is smaller and no total cancelling can occur, as shown in curve b).
What will be the envelope, if the load impedance is the same as the characteristic impedance
of the transmission line?
(Equation 54) results in the following relation between the phase angle of the reection co-
ecient φ at the point z =0 and the distance between z =0 and the rst minimum voltage
(n=0):
φ = −2β∆z − π . (57)
From (42), (43), (44), and (48), we can derive the below relation:
1 − m −j(2β∆z+π)
ZL 1+ e
= 1 +m . (58)
Z0 1 − m −j(2β∆z+π)
1− e
1+m
From this it results after some conversions:
ZL m + j tan(β∆z)
= . (59)
Z0 1 + j m tan(β∆z)
The unknown impedance Z L can be dened in the case the following three parameters are
known: m from (42), the wave impedance of the transmission line Z0 , and the relative position
3 Experiment set-up
On page 14 Fig. 7 you nd the schematic test arrangement. The 3 GHz oscillator provides
the basic frequency for the measurement. The frequency is amplitude modulated by a 1 kHz
oscillator and a PIN diode switch. The isolator and the attenuators protect the components
from reected waves. The probe detects the distribution of the voltage on the measuring trans-
mission line and by means of the detector diode and the bandpass lter the 1 kHz modulated
frequency is ltered out, amplied and nally measured.
The modulation is not really necessary, but an alternating voltage of this low frequency is
much easier to handle than a direct voltage.
The transmission line used for measurement consists of two parallel plates arranged on the same
potential and of a round internal conductor. In Fig. 6 the scheme for measuring transmission
line without sockets and the probe in comparision with a coaxial line are shown.
a) b)
Problem 1: Write down the expression for a forward wave and a backward wave separately.
Problem 2: Two wires several wavelengths long serve as a connection between a generator
and a load. The distance between the wires is small but not constant, varying as a
smooth function along the line.
Can you use the transmission line equation (17) for the analysis of this transmission
line?
λ λ
Problem 3: What are the input impedances of lossless transmission lines of lengths
4 and 2
in case they are short-circuited or open-ended?
Problem 4: What is the eect on the graph of the voltage in Fig. 5 a), in the case the
Problem 5: How many minima and maxima are detectable if a transmission line of length
Problem 7: What is more useful for the denition of the wavelength; to measure the minima
5 Measuring Tasks
Make sure that the experimental set-up is build according to Fig. 7. The RF components
and the golden SMA connections are damageable and expensive, therefore, please
don't dismantle the connections. On the voltage meter the unit is indicated as µA.
Anyway it is a voltage meter controlled by the amplied and ltered diode signal. Since only
the relative values are of interest, please take a.u. (arbitrary unit) as unit. Pay attention to
the algebraic sign of the measured distances!
Useful formulas:
m + j tan(β∆z)
Z= Z
1 + jm tan(β∆z) 0
Z0 = 50 Ω
∆z =zmax − zSC
zmax = Position of the measured maximum
zSC = Position of the rst measured maximum at the calibration with a short circuit
Umin
m=
Umax
Calibration:
Unscrew the probe slightly with the knurled screw. Connect a short circuit to the transmission
line; move the measuring slide to the global maximum and revolve the probe till the voltmeter
shows full-scale deection. After this calibration the depth of penetration of the probe must
not be changed! Not the position of the rst maximum zSC .
Determine the locations and the appropriate amplitude of the rst two maxima and
minima.
Measure now without a new calibration the locations and amplitudes for the case the
transmission line is open-ended.
Why do the amplitudes of the maxima dier?
Task 2: Determine as exactly as possible the wavelength on the transmission line using the
results from the previous task and from this value the frequency of the RF source. The
frequency of the RF source of set-up 1 is 2.999 GHz, that one of set-up 2 3.003 GHz.
Calculate the relative error.
Task 3: Connect the 50 Ω N-connector impedance (this is the value of the characteristic
impedance of the transmission line) to the transmission line and determine the standing
wave ratio SWR. Calculate the amplitude of the reection coecient in dB. What is
theoretical value for the standing wave ratio and the reection coecient in this case?
What are the reasons for the dierence between theory and practice?
Task 4: The impedance Z2 (marked with 4 dots in the brass bulk) consists of a metal lm
resistor of 100 Ω which acts only at low frequencies (below MHz range) as wanted.
Determine the complex impedance and from this a simple equivalent circuit consisting
of R and L or C for the given frequency. Determine the values of the equivalent circuit.
Task 5: The impedance Z1 (marked with one dot in the brass bulk) consists of a short
open-ended coaxial transmission line, lled with the dielectric media with εr =2.0. The
characteristic impedance of the coaxial transmission line is 50 Ω as of the air-lled trans-
mission line.
Determine the complex impedance of the circuit and from this the length of the trans-
mission line.
Hint: arccot(x) = π
2 − arctan(x).
Task 6: Calibrate the transmission line using an SMA short circuit and an N-SMA transition.
Task 8: Determine the values of SWR as well as the magnitude of the reection factor in dB
for the four congurations of SMD resistors (1×50 Ω, 2×100 Ω in parallel, 3×150 Ω in
parallel and 4×200 Ω in parallel).
Calculate the theoretical value of the reection factor in dB for the case the load resistor
has 55 Ω.
Why the measured results for the resistor congurations dier?
Task 9: Where are the limits of the measuring method and what errors exist? Why the
measurement results are not ideal?
Oscillator
Detector diode
1 kHz
3 GHz
3 dB Probe
α
Measuring
13
5 Measuring Tasks
5 Measuring Tasks